1. Field of the Invention
The present invention generally relates to transmission line-to-waveguide transitions, and more specifically to an end-on transition that does not contact the waveguide when inserted.
2. Description of the Related Art
Transmission line-to-waveguide transitions are used extensively in microwave communications systems such as radar and satellite systems. The systems may include a waveguide antenna for phased array applications or a conventional waveguide of arbitrary cross-section. In these systems the microwave signal is ideally bi-directionally coupled between a waveguide and a transmission line with minimal power (insertion) loss and maximum signal clarity. The transmission line can be a monolithic microwave integrated circuit (MMIC) that is wire bonded to the transition or it can be a coaxial cable.
A major source of loss in microwave systems is impedance mismatch between components. The mismatch causes a significant portion of the signal to be reflected at their junction. Therefore, matching the impedances of the components is very important for reducing the transition's insertion loss.
A common end-on transition between a coaxial cable and a waveguide is described by Deshpande, "Analysis of an End Launcher for an X-Band Rectangular Waveguide", IEEE Transactions on Microwave Theory and Techniques, Vol MTT-27, No. 8, August 1979, pp. 731-735. The transition is formed by bending the cable into an L-shaped loop, grounding its outer conductor and attaching (welding) its center conductor to the waveguide. The direct contact between the transition and the waveguide makes the transition's impedance difficult to calculate. It is difficult to design the dimensions of the loop to provide a wide bandwidth with low insertion loss while maintaining tight enough manufacturing tolerances to achieve the designed bandwidth. Furthermore, forming a high quality contact between the coax and waveguide adds substantially to the manufacturing cost of the transition.
Another type of transition is the antipodal finline disclosed by Ponchak, "A New Model for Broadband Waveguide-to-Microstrip Transition Design", Microwave Journal, May 1988, pp. 333-343. In this transition, finline conductors on opposing sides of a substrate form a high quality contact with the waveguide's inner walls. As a result, the conductors require unusual and complicated cross sectional designs to efficiently couple the signals between the waveguide and the microstrip. A semicircular fin is positioned next to one of the finlines to adjust the resonant frequency of the transition.
An external dipole transition to a ridge waveguide is disclosed in U.S. Pat. No. 5,095,292 to Masterton. The dipole coupling must be at least 0.5 wavelengths in size and is restricted to ridge waveguides. The dipole coupling is prohibitively large for phased antenna arrays, which typically have center-to-center spacings less than 0.5 wavelengths. U.S. Pat. No. 4,905,013 to Reindel describes a finline horn antenna that includes a finline dipole radiator extending a quarter-wave out from an open ended waveguide. The finline slot forms a high quality contact with the waveguide. U.S. Pat. No. 4,425,549 to Schwartz discloses conductive finlines disposed on opposite surfaces of a dielectric substrate and in direct contact with the inner walls of a rectangular waveguide. A diode that connects the opposing finlines is used to couple RF signals in the waveguide to a filter. A balun for directly coupling microwave signals between a spiral antenna and a transmission line is described by Bawer, "A Printed Circuit Balun for Use with Spiral Antennas", IRE Transactions on Microwave Theory and Techniques, May 1960, pp. 319-325.
In all of the transmission line-to-waveguide transitions except Masterton's external dipole, the slotline or finline is permanently attached to the inner walls of the waveguide to form a high quality mechanical and electrical contact. Welding the transition directly to the waveguide is a difficult and expensive process. It is difficult to manufacture the contact with the tight tolerances and quality required to achieve a large bandwidth with low insertion loss.
The transition and waveguide which it contacts form a three dimensional system that is very difficult to model, one reason being that the charge density does not uniformly decrease away from the slotline's inner surfaces. Instead the charge tends to accumulate at the transition-waveguide contacts, which greatly increases the complexity of the impedance computations. Furthermore, the bandwidth (10%-15% of the center frequency), insertion losses and the tuning of the resonant frequencies in the waveguide are not optimum.